Into the Fray: Cardioversion Without OAC in Watchman Patients

Patrice Wendling

October 30, 2019

A small series provides provocative new data on the safety of performing direct current cardioversion (DCCV) without oral anticoagulation in high-risk patients with a Watchman left atrial appendage occlusion (LAAO) device.

There are no guidelines regarding DCCV in patients with LAAO devices, despite more than 100,000 Watchman devices implanted worldwide.

To investigate, Sharan Prakash Sharma, MD, from the Kansas City Heart Rhythm Institute and Research Foundation in Overland, Kansas, and colleagues examined outcomes retrospectively in 148 consecutive patients with atrial fibrillation (AF) or atrial flutter who had DCCV 1 week to more than 1 year after Watchman implantation (median, 5.1 months) at multiple sites.

Their mean CHA₂DS₂-VASc and HAS-BLED scores were 3.8 and 3.4, respectively. AF was paroxysmal in 49%, persistent in 44%, and long-standing persistent in 7%.

Transesophageal echocardiography (TEE) performed in all patients just prior to DCCV revealed device-related thrombus in four patients, who were subsequently treated with a direct oral anticoagulant (DOAC) and underwent cardioversion after resolution of the thrombus on repeat imaging.

Importantly, none of the patients had a peridevice leak of 5 mm or greater, incomplete device apposition, or embolization on TEE prior to cardioversion, the authors note.

DCCV restored sinus rhythm in all patients and no patient developed DCCV-related thromboembolic complications. Three transient ischemic attacks (TIAs) occurred but were not thought related to DCCV as they occurred 3 to 4 months after cardioversion. All three patients were taking aspirin before the procedure.

"The old thought was when you do a cardioversion in these people in the absence of anticoagulation and a left atrial appendage occlusion device, you had a significant amount of stunning that happens inside the left atrial appendage and you will put them at risk for thrombus formation in the immediate aftermath of doing the cardioversion," said senior author Dhanunjaya Lakkireddy, MD, also with the Kansas City Heart Rhythm Institute.

"But what we were able to appreciate is even when they're not anticoagulated afterwards, their risk of stroke is actually extremely low to nonexistent," he told theheart.org | Medscape Cardiology.

US guidelines recommend anticoagulation in patients with AF for at least 3 weeks before and for at least 4 weeks after DCCV. They also say it's reasonable to proceed with cardioversion without prior anticoagulation if no LAA thrombus is present on TEE and postprocedure anticoagulation is maintained for at least 4 weeks.

At the time of cardioversion, 34% of patients were on oral anticoagulation, 30% dual antiplatelet therapy (aspirin 81 mg daily and clopidogrel 75 mg daily), and 36% on aspirin 81 mg. Immediately after DCCV, 22% were newly started on oral anticoagulation (median duration, 12 weeks).

"Individual physicians treated them differently and that, in a way, was very helpful for us because it actually helped us to figure out the impact of using just oral antiplatelet agents or using oral anticoagulation for a month or not using any anticoagulation and antiplatelets," Lakkireddy said.

Bleeding events occurred in 4% of patients and were minor in two and major in four (three GI bleeds requiring transfusion and one intracranial hemorrhage).

Although bleeding events trended higher in those treated with oral anticoagulation after DCCV (6.0% vs 1.5%; = .20), there were no major differences in age or CHA₂DS₂-VASc and HAS-BLED scores between patients with and without these events, the authors reported October 28 in the Journal of the American College of Cardiology.

In follow-up up to 12.8 months, no device or left atrial thrombosis, device dislodgement, or new device leak were reported. One patient died from cancer.

In a related editorial, John Mandrola, MD, Baptist Health, Louisville, Kentucky, and colleagues say they welcome the new data, given the "near absence" of evidence in this area, but also caution they are preliminary data in a small number of patients. By comparison, the Canadian AF treatment guidelines sampled thousands of patients for its 1-month estimates of stroke or systemic embolism after DCCV of 0.46% for patients treated with warfarin and 0.31% for those taking DOACs.

"Given the low rate of events after DCCV, the lack of events in a series of 148 patients could be due to chance," they say. "If just 2 post-procedure events occurred in the next 50 patients, the rate would be more than double the established safety threshold."

The editorialists also take aim at the argument that LAAO simplifies therapy by removing anticoagulation, noting that all patients had at least one TEE, 58% were on oral anticoagulation, and some patients were also maintained on aspirin despite randomized data showing apixaban to be superior to aspirin, with no difference in major or intracranial bleeds.

Addressing the latter point, Lakkireddy commented: "These are the ones where the operator truly believed that the risk of putting them back on oral anticoagulation was significantly higher than not using any DOAC, even for a short term, making a point that oral anticoagulation in these cases, in some of these very tenuous cases, may be very difficult. At the same time, these are the same people that also needed a cardioversion because of their symptoms, so rhythm control was important in these patients."

The overall results should be taken with "a pinch of salt because our N is relatively small," he said, but that it provides a "good signal to get started on."

In fact, the team is writing a protocol for a randomized controlled trial at 15 to 20 sites looking at this issue in 400 to 500 all-comers with LAAO devices.

"I think it's safe to say that it is feasible, but we are not claiming anything more than that," Lakkireddy said. "As long as people do a preprocedural TEE, I think we are in very good shape."

Finally, Mandrola and colleagues express concern about the technology as a whole and that the study likely will be cited as showing the feasibility of DCCV after LAAO.

"Embedded in these citations is the implied and hidden idea…well, if they (the experts) are doing cardioversion after appendage occlusion, and journals are publishing safety data, then appendage occlusion must be reasonable," they write. "We strongly push back on this implication."

They remind readers that the PROTECT-AF study of the Watchman device "did not pass FDA muster" and that the subsequently mandated PREVAIL study missed noninferiority for its first coprimary end point and barely met its second coprimary end point.

Asked whether he felt the data establish the feasibility of DCCV after LAAO, Lakkireddy said: "I would be the biggest critic of my own paper. I don't think we've proven anything here in a big way but at least we have a very good signal to basically look at this as a concept and at least it feels safer when this dialogue comes up every single time."

"After we presented this abstract at Heart Rhythm Society last year, you can't believe how many electrophysiologists and interventionalists called me from all over the country asking me this very question. I used to be one of those guys who asked this question all the time myself, and that's where the study came from."

"It's easy for people who sit in the pulpit to throw a stone at evidence that is just being generated and create unnecessary anxiety," he added. "People can be cautious but yet optimistic. I think that is the message that one has to realize. That is the bottom line offset. So I wouldn't claim too many things, but if we don't generate science, who else is going to push it forward? That is the attitude we need."

Lakkireddy has served as a consultant to Boston Scientific, Biotronik, Medtronic, and Biosense Webster. Mandrola is a regular contributor to Medscape.

J Am Coll Cardiol. 2019;74:2267-2274 and 2275-2277. Abstract, Editorial

Follow Patrice Wendling on Twitter: @pwendl. For more from theheart.org | Medscape Cardiology, join us on Twitter and Facebook.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....